Circuit-Forge/turnstone
5
0
Fork 0
Code Issues 11 Pull requests 1 Projects Releases 6 Packages Wiki Activity Actions

docs(container): fix GPU_SERVER_URL for Xander — use public orch.circuitforge.tech

Browse source 
Xander's example-node.tv has no WireGuard route to Heimdall's LAN (10.1.10.x),
so the <YOUR_HOST_IP>:7700 private address is unreachable from there.

Use the public cf-orch endpoint instead:
  GPU_SERVER_URL=https://orch.circuitforge.tech

Daniel's Huginn has WireGuard to Heimdall LAN — <YOUR_HOST_IP>:7700 stays correct.
Added both options to docker-standalone.sh for clarity.
This commit is contained in:
pyr0ball 2026-05-26 13:39:38 -07:00
parent d57d44e285
commit 0d43b23f0c
2 changed files with 12 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
docker-standalone.sh
View file

@ -85,13 +85,17 @@ TZ="${TZ:-America/Los_Angeles}"
# ── Multi-agent diagnose pipeline ──────────────────────────────────────────── # ── Multi-agent diagnose pipeline ────────────────────────────────────────────
# Enable the 5-stage ML pipeline to get smarter diagnose results. # Enable the 5-stage ML pipeline to get smarter diagnose results.
# Requires WireGuard connectivity to Heimdall (<YOUR_HOST_IP>) for LLM synthesis.
# #
# For Daniel's instance (Huginn) — Heimdall's cf-orch via WireGuard: # If your host has WireGuard to Heimdall's LAN (e.g. Huginn):
# export GPU_SERVER_URL=http://<YOUR_HOST_IP>:7700 # export GPU_SERVER_URL=http://<YOUR_HOST_IP>:7700
# export TURNSTONE_MULTI_AGENT_DIAGNOSE=true # export TURNSTONE_MULTI_AGENT_DIAGNOSE=true
# bash ~/turnstone/docker-standalone.sh # bash ~/turnstone/docker-standalone.sh
# #
# If your host has no WireGuard to Heimdall (use public cf-orch endpoint):
# export GPU_SERVER_URL=https://orch.circuitforge.tech
# export TURNSTONE_MULTI_AGENT_DIAGNOSE=true
# bash ~/turnstone/docker-standalone.sh
#
# ML models are downloaded on first diagnose run and cached in HF_CACHE_DIR. # ML models are downloaded on first diagnose run and cached in HF_CACHE_DIR.
# First run takes a few minutes (downloading ~400MB of CPU-only models). # First run takes a few minutes (downloading ~400MB of CPU-only models).
# Subsequent runs are instant (models served from hf-cache/). # Subsequent runs are instant (models served from hf-cache/).

10
podman-standalone.sh
View file

@ -91,15 +91,17 @@ TZ=America/Los_Angeles
# ML models are downloaded on first diagnose run and cached in HF_CACHE_DIR. # ML models are downloaded on first diagnose run and cached in HF_CACHE_DIR.
# On a CPU-only host (no GPU) set TURNSTONE_EMBED_DEVICE=cpu (default). # On a CPU-only host (no GPU) set TURNSTONE_EMBED_DEVICE=cpu (default).
# #
# For Xander's instance (example-node.tv) — Heimdall's cf-orch via WireGuard: # For Xander's instance (example-node.tv) — no WireGuard to Heimdall LAN,
# export GPU_SERVER_URL=http://<YOUR_HOST_IP>:7700 # use the public cf-orch endpoint instead:
# export GPU_SERVER_URL=https://orch.circuitforge.tech
# export TURNSTONE_MULTI_AGENT_DIAGNOSE=true # export TURNSTONE_MULTI_AGENT_DIAGNOSE=true
# sudo bash /opt/turnstone/podman-standalone.sh # sudo bash /opt/turnstone/podman-standalone.sh
# #
# For Daniel's instance (Huginn) — same cf-orch via WireGuard: # For Daniel's instance (Huginn) — WireGuard reaches Heimdall LAN directly,
# use docker-standalone.sh (not this script — Docker host):
# export GPU_SERVER_URL=http://<YOUR_HOST_IP>:7700 # export GPU_SERVER_URL=http://<YOUR_HOST_IP>:7700
# export TURNSTONE_MULTI_AGENT_DIAGNOSE=true # export TURNSTONE_MULTI_AGENT_DIAGNOSE=true
# bash /opt/turnstone/podman-standalone.sh (or rebuild Docker container) # bash ~/turnstone/docker-standalone.sh
# ── Turnstone container ─────────────────────────────────────────────────────── # ── Turnstone container ───────────────────────────────────────────────────────
# Image is built locally — no registry auto-update label. # Image is built locally — no registry auto-update label.